Hi everyone,
I’m running into a strange issue with the Gemini AI integration (I'm using the TextFromImage Gemini integration built by Flutterflow) in FlutterFlow and I wonder if anyone else has faced (and solved) this.
When I first set up Gemini with no Primary Language configured, everything worked perfectly.
As soon as I set Primary Language = Portuguese, I started getting this error:
gemini-1-5-flash-001-tuning-is-not-found-for-api-version-v1beta
I deleted and re-added the integration, but the issue continued.
When I switched Primary Language to English, I got a different error:
publisher model projects/generativelanguage-ga/locations...
It seems that when a Primary Language is set, FlutterFlow builds the wrong model path for Gemini.
From Google’s docs, the only valid model paths should be:
publishers/google/models/gemini-1.5-flash
publishers/google/models/gemini-1.5-pro
…but FlutterFlow is generating projects/generativelanguage-ga/...
or appending -tuning
, which don’t exist.
👉 Expected behavior: I’m trying to use Gemini to extract text from images in Portuguese. It should just call the valid model path and return text normally.
Has anyone found a workaround?
Any help would be really appreciated 🙏
Thanks,
Ewerton